Kinetics of symmetric proton transfer reactions in alcohols and amines

Apr 1, 1977 - Experimental Determination of Activation Energies for Gas-Phase Ethyl and n-Propyl Cation Transfer Reactions. Travis D. Fridgen and Terr...
0 downloads 0 Views 659KB Size
T H E

J O U R N A L

O F

PHYSICAL CHEMISTRY Registered i n U. S. Patent Office

0 Copyright, 1977, by the American Chemical Society

VOLUME 81, NUMBER 7 APRIL 7, 1977

Kinetics of Symmetric Proton Transfer Reactions in Alcohols and Amines by Ion Cyclotron Resonance Spectroscopy. Relaxation Mechanisms for Epithermal Ion Energy Distributions 1.B. McMahon' and J. L. Beauchamp*2a Arthur Amos Noyes Laboratory of Chemical Physics:b California Institute of Technology, Pasadena, California 9 1 125 (Received January 3, 1977)

Rates of symmetric proton transfer have been measured for a number of simple aliphatic alcohols and amines both by the conventional ion cyclotron resonance intensity vs. pressure methods and using trapped-ion techniques. For the systems studied symmetric proton transfer was found to be very fast, occurring essentially at the ion-molecule encounter rate. The implications of the speed of these reactions with respect to the relaxation of epithermal ion energy distributions is discussed.

excess internal and translational excitation will not be relaxed prior to reaction. Hence most observed processes are not representative of thermal ion energy distributions. Failure to thermalize reactant ions is of particular concern in cases where thermochemical inferences are made from the preferred direction of reaction or in favorable cases from equilibrium constants for reversible bimolecular ion-molecule reaction^.^ This concern is justified on the basis of disagreements which exist in the literature for thermochemical quantities determined by different investigators using different experimental conditions and, possibly, different ion energy distributions. It is thus important to determine the conditions which promote rapid thermalization of gas phase ions. General agreement among various research groups using different experimental techniques has been obtained for alkylamine basicities! Thislends support to the conjecture that highly accurate thermochemical data may be derived from gas phase proton transfer equilibria measurements of the type generalized by

Introduction Numerous factors can frequently lead to production of nonthermal energy distributions for gas phase ions. A substantial fraction of ions formed by electron impact or photoionization above the adiabatic ionization threshold will often possess considerable internal excitation. In some cases ions may be subjected to accelerating fields which lead to varying degrees of translational excitation dependent on the nature of the experimental apparatus. In formation of fragment ions considerable translational energy may be released from dissociation of the excited parent ion, particularly if the ionization occurs by a transition to a repulsive electronic state of the ion. In addition, product ions resulting from exothermic ionmolecule reactions may possess a substantial fraction of the exothermicity of reaction as internal and translational excitation. Haney and Franklin3 have recently examined the partitioning of excess energy between translational and internal degrees of freedom for both fragmentation processes and ion-molecule reactions and have found that vibrational excitation of fragment and product ions may be appreciable. Experimental observations have shown that most ionmolecule reactions are efficient, with reaction often taking place on every collision which brings the reactants in close contact. Thus, in the absence of a nonreactive diluent gas

B,H+ + B, 2 [B,HB;]* 2 B,

+ B,H+

(1)

where B1 and B2represent n-donor bases. It has recently been demonstrated that symmetric proton transfer reactions of the form of eq 1where B1and B2 are identical, may be an efficient means of relaxing excited ions! 593

594

T. 6.McMahon and J. L. Beauchamp

Considerable experimental evidence suggests that proton transfer reactions proceed via an intermediate proton bound dimer as indicated in eq L7 This intermediate may dissociate with the labile proton bound to either of the n-donor bases involved. If the intermediate has a lifetime which is short relative to the time scale for redistribution of vibrational energy then the neutral product, B1, for reaction 1proceeding to the right, will carry off the bulk of the excess energy in a single proton transfer event. If the lifetime of the proton bound dimer is long compared to the time for redistribution of vibrational energy then dissociation of the intermediate will likely involve partitioning of the excess energy between the ionic and neutral fragments based on statistical considerations involving the relative numbers of degrees of freedom in the products. In either case it is apparent that symmetric proton transfer reactions in which an n-donor base interacts with its own conjugate acid will provide an efficient means of thermalizing reactant ions if such processes are fast. For a partially deuterated species MH, where n hydrogens are directly bonded to the n-donor site it is often observed that the parent ion MH,' reacts with the parent neutral to form MH,D+. This species in turn undergoes the reaction MH,D+ t MH,

+

MH,+{

+ MH,-lD

(2 1

resulting in the disappearance of labile deuterium in the conjugate acid. For example, methanol-d3, CD,OH, undergoes reactions 3 and 4 with the product of reaction 4 CD,OH+ + CD,OH

-r

CD,OH:

+ CD,O

CD,OHD+ + CD,OH CD,OHD+ t CD,OH + CD,OH,i t CD,OD --+

CD,OH,i t CD,OH

--f

(CD,),OH+ + H,O

(3) (4) (5 1 (6)

subsequently being consumed in reaction 5. Sieck, Abramson, and Futrel18have noted that labile deuterium does not appear in a product of condensation reaction such as reaction 6 indicating that the symmetric proton transfer reaction 5 is fast compared to protonated dimethyl ether formation (6). Munsong has also studied the generalized process 2 for deuterated ammonium ions. Reaction 7, in which NH3D+ NH,D+ + NH,- NH,'

+

NH,D

(7 1

is formed by deuteron transfer employing CD4 as a chemical ionization reagent, was found to be relatively fast (k7 = 1X lo4 cm3molecule-' d).This compares favorably with our own value of k8 = 1.2 X lo-' cm3 molecule-' for reaction '8 determined using trapped-ion ICR ion 14NH,' t 15NH,+ "NHqC t 14NH,

(8)

ejection techniques. Related and more extensive studies have been reported by Harrison and co-workers." Using CD4 as a reagent gas they examined symmetric proton transfer reactions involving NH3, CHSNHZ, (CH3),NH, CD30H, and CH3SH. In the present study we report kinetic data for reactions analogous to reaction 5 for specifically labeled methanol, ethanol, methylamine, and dimethylamine as well as reactions and rate constants for accompanying processes. The site specificity of processes analogous to reactions 3 and 4 is analyzed for the first time in several of the systems studied. Rate constants are derived both by consideration of the variation of ion abundance with pressure" and the variation of ion abundance with time at fixed pressure in trapped-ion ICR experiment~.'~ The results are discussed in terms of models for proton transfer and the implications The Journal of Physical Chemistry, Val. 8 1, No. 7, 1977

i

1"0-6

IO-' Pressure ( t o r r )

Figure 1. Variation of ion abundance with pressure for CD3NH2at 11.0 eV. The solid line indicates the variation of ion intensity with pressure calculated from the average rate constant obtained from several determinations. The points indicate experimentally determined intensities.

arising for the deexcitation of protonated n-donor bases are considered.

Experimental Section All experiments in this study were carried out at ambient temperature (23 "C) on a Varian lnstruments V-5900 ICR spectrometer modified to permit trapped-ion experiments to be carried out routinely. Pressures below low5Torr were monitored using an ionization gauge of the Shulz-Phelps design.14 Above Torr absolute pressures were measured by an MKS Instruments Baratron capacitance manometer which also permitted calibration of the ionization gauge. Rate constants from trapped-ion spectra were obtained from semilogarithmic plots of relative ion abundance vs. time. Rate constants from intensity vs. pressure data were calculated by iterative computer solution to the ICR power absorption equations formulated The Marshall-Buttrill by Marshall and Buttrill." equations treat the case of a single primary ion reacting to produce several secondary ions which each may subsequently react to form different tertiary ions. Complications arise in the kinetic analysis when an ion may be both a secondary and a tertiary ion (for example, CD30H2+ in reactions 3 and 5). In this, and related cases, the different contributions to CD30H2+were analyzed separately in computations of rate constants but were summed to compare calculated to observed ion intensities. The set of rate constants reported for each system are those which give the best fit of calculated to observed intensity vs. pressure data. Methylamine-d3 and dimethylamine-de were obtained from Merck Sharpe and Dohme of Canada Ltd. Methanol-d3and ethanol-d5were prepared by refluxing the fully deuterated analogues, obtained from Merck Sharpe and Dohme of Canada Ltd., with magnesium turnings to produce the dialkoxy magnesium salts. These salts were subsequently treated with HZOand distilled to obtain the exchanged alcohols. All samples were routinely degassed by bulb-to-bulb distillation. Analysis by mass spectrometry indicated the isotopic purity to be greater than 99 atom % D at the labeled sites. Results 1. Methylamine. At sufficiently low electron energies only the parent ion of methylamine is formed. In the case of CDBNHzthis species undergoes reactions 9 and 10 CD,NH,'

CD,NH;

+ CD,NH,

-I-'

+ CD,NH

+CD,NH,D+

+ CD,NH,

(9) (10)

producing CD3NH3+and CD3NHzD+,respectively. It can be seen from the variation of ion abundance with both

595

Proton Transfer Reactions in Alcohols and Amines

\

\

40

80

120

160

200

Time (rnsec)

Figure 2. Variation of ion abundance with time for CD3NHzat 12.0 eV and 6.5 X lo-' Torr with a 6-ms electron beam pulse.

Flgure 4. Variation of ion abundance with time for (CD&NH at 10.5 Torr with a 6-ms electron beam pulse. eV and 7.5 X

-

Pressure (torr)

Pressure (torr)

Figure 3. Variation of ion abundance with pressure for (CD&NH at 11.O eV. The solid line indicates the variation of ion intensity with pressure calculated from the average rate constant obtained from several determinations. The points indicate experimentallydetermined intensities.

pressure (Figure 1) and time (Figure 2) that the ratio of CD3NH3' to CD3NH2D+is not constant as the reaction sequence progresses. The deuterated species is converted to the protonated ion by the symmetric proton transfer reaction CD,NH,D+

+ CD,NH,

+

CD,NH,'

+ CD,NHD

(11)

Rate constants for reactions 9-11 as well as the rate constant for formation of CH3NH3+from CH3NH2are summarized in Table I. Included for comparison are available literature data for these reactions. 2. Dimethylamine. Reactions 12-14 are observed for WD,),"' --i

+ (CD3),"+

(CD3)2"2+ (CD,),NHD+

(CD,),NHD+

+

(CD,),NH

-t

(CD,),",'

1. i

+ (CD,)zN

(12)

+ CD,NHCD, I (13) + (CD,),ND

(14)

(CDJ2NH in direct analogy with the related processes

Flgure 5. Variation of ion abundance with pressure for CDBOHat 11.O eV. The solid line indicates the variation of ion intensity with pressure calculated from the average rate constant obtained from several determinations. The points indicate experimentally determined intensities.

occurring in methylamine. Rate constants for reactions 12-14 as determined from the variation of ion abundance with pressure (Figure 3) and time (Figure 4) are summarized in Table I together with data for the unlabelled species. 3. Methanol. The ion-molecule reactions of simple alcohols are slightly more complex than those for alkylamines due to formation of protonated dialkyl ethers by nucleophilic displacement reactions.15-17 The sequence of reactions 3-6 is apparent in the variation of ion abundance with pressure (Figure 5) and time (Figure 6). In the trapped-ion experiment it was necessary to operate at slightly higher electron energy resulting in the production of CD20H+as well as the parent ion. This species reacts exclusively by proton transfer to form CD30H2+.Proton bound dimer formation is also observed at pressures above Torr (Figure 5). The absence of the proton bound dimer in the trapped ion data of Figure 6 indicates that collisional stabilization of the excited intermediate formed in the reaction of CD30H2+with CD30H is required to observe this species. Rate constants for ion-molecule The Journal of Physical Chemistv, Vol. 8 I, No. 7, 1977

596

T. 6. McMahon and J.

L. Beauchamp

TABLE I: Rate Constants for Ion-Molecule Reactions of Simple Alcohols and Amines Rate constant This work Intensity vs. pressure

Intensity vs. time

1. Methylamine CH,NH,+ + CH,NH, + CH,NH,+ + CH,N CD,NH,+ t CD,NH, + CD,NH,+ t CD,NH CD,NH,+ t CD,NH, CD,NH,D+ + CD,NH, CD,NH,D+ + CD,NH, CD,NH,+ t CD,NHD

20.6 12*2 5.8 4.8

}

16.4 18.9

2. Dimethylamine (CH,),NH,+ t C,H,N (CH,),NH+ + (CH,),NH (CD,),NH+ + (CD,),NH + (CD,),NH,+ + (CD,),N (CD,),NH+ + (CD,),NH (CD,),NHD+ t CD,NH(CD,) (CD,),NHD' + (CD,),NH (CD,),NH,+ t (CD,),ND

14.9 7'5 7.0 3.4

1

3. Methanol CH,OH' t CH,OH- CH,OH,' t CH,O CH,OH,' t CH,OH -+ (CH,),OH+ t H,O CD,OH+ t CD,OH-+ CD,OH,+ t CD,D CD,OH+ t CD,OH-+ CD,OHD+ t CD,OH CD,OHD' t CD,OH + CD,OH,+ + CD,OD C D 3 0 H z +t CD,OH- (CD,),OH,+ t H,O

21.7 1.0 12*7 14.6 4.4 1.5

Reaction

-+

-+

--f

-+

4. Ethanol CH,CH,OH+ t CH,CH,OH CH,CH,OH,+ + C,H,O CH,CHOH+ + CH,CH,OH -+ CH,CH,OH,+ + CH,CHO CH,CH,OH,+ t CH,CH,OH + (CH,CH,),OH+ + H,O CD,CD,OH' t CD,CD,OH -+ CD,CD,OH,+ + CD,CD,O CD,CD,OHD+ t CD,CDOH CD,CD,OH+ t CD,CD,OH CD,CD,OH,+ + CD,CDO CD,CDOH' + CD,CD,OH CD,CD,OHD+ t CD,CD,OH CD,CD,OH,+ + CD,%U,OD CD,CD,OH,+ + CD,CD,OH + (CD,CD,),OH+ t H,O

3.1

2.8

1

-2.2

}

-+

-+

' M. S. B. Munson, J. Phys. Chem., 70, 2034 (1966).

- 3.7 -1.7

20.8 1.1 24.0 4.2 1.3

--f

-+

12.8 13.5

17.8 11.2 2.4 13.6 10.4 2.8 1.4

(X

10" cm3 molecule-' s-l)

Other 21.0,' 9 6.4

*

l,*16,c 12.2,d 12.1d

I0.2e

12: 6

11.6:

Il,*

10.2,d 10.2d

3.1 * 0.2f 25.3g

go:;

4.0

* 0.5h

12.0'

*

E. G. Jones and A. G. Harrison, Can. J. Chem., 45,3119 (1967). 39,6 3 (1965). L. Hellner and L. W. Sieck, Int. J . Chem. Kinet., 5, 177 (1973). M. J. Henchman, Discuss. Faraday SOC., e Reference 1 0 (rate constant reported for CH,NH,D+). f Reference 10 (rate constant reported for (CH,),NHD+). g S. K. Gupta, E. G. Jones, A. G. Harrison, and J. J. Myher, Can. J. Chem., 45, 3107 (1967). Reference 1 0 . i L. W. Sieck, F. P. Abramson, and J. H. Futrell, J. Chem. Phys., 45, 1655 (1966).

CD,CD,OH;

I

1

10-3

Pressure (torr)

Figure 7. Variation of ion abundance with pressure for CD3CD,0H at

12.0 eV.

T i m e (rnsec)

Figure 6. Variation of ion abundance with time for CDBOHat 14.0 eV and 3.5 X IO-' Torr with a 6-ms electron beam pulscL

reactions in CH30H and CD30H are summarized in Table I. Rate constants for symmetric proton transfer (reaction The Journal of Physical Chemistv, Vol 8 I, No. 7, 1977

5 ) and protonated dimethyl ether formation (reaction 6) are derived from an analysis of the kinetics of disappearance of the deuterated and protonated parent molecules, respectively. 4. Ethanol. In addition to the parent ion of ethanol, it was found necessary to operate even in the pressure variation experiment (Figure 7) with the a-cleavage fragment, CH3CHOH+,present. A t the slightly higher electron energy used in the trapped ion experiment (Figure 8) the fragment CH20H+also appears. In both cases the fragment ions are protonated aldehydes and react exclusively via proton transfer to form the protonated parent ion. The complications introduced by the presence of CH3CHOH+in the data of Figure 7 made application of the Marshall-Buttrill analysis difficult since two primary ions react to produce the same secondary ion. However rate constants for reaction of all three primary ions in the data of Figure 8 are easily determined due to the greater

597

Proton Transfer Reactions in Alcohols and Amines

TABLE 11: Comparison of Parent Ion and Proton Transfer Rate Constants Statistically Total corrected reaction proton rate of transfer parent encounter ADOapC Lange. Soecies iona ratea>l' theory vin

~

~

~~

15.5d 18.5 14.0 5.lf 25.6 13.6

3 "

CD,NH, (CD,),NH (CD3)J CD,OH CD,CD,OH

12.0e 11.9 12.4 5.5f 17.2 13.2

21.5 15.9 13.6 10.9 17.8 16.2

11.6

11.1

10.9 10.9 10.0 10.3 See a All rate constants x 1O1O cm3 molecule" s" . text. Reference 10. W. T. Huntress, Jr., and R. F. Pinizzotto, Jr., J. Chem. Phys., 59, 4742 (1973). e Ref. erence 11. Reference 20.

of either a deuteron bound dimer, I, or proton bound R -0

I \

0.02 -

I

R--0 I Time (msec)

Figure 8. Variation of ion abundance with time for CD3CD,0H at 16.0 eV and 8.0 X lo-' Torr with a 6-ms electron beam pulse.

ease of kinetic analysis by the trapped ion method. The ion chemistry initiated by the parent ion of ethanol is described by the sequence of reactions 15-18 analagous CD,CD,OH+

+ CD,CD,OH

CD,CD,OH,+

+ CD3CD,0

(15)

CD,CD,OHD'

+ C,D,OH

(16)

-

CD,CD,OHD' + CD,CD,OH CD,CD,OH,' + CD3CD,0H-

CD,CD,OH,'+

CD,CD,OD

(CD,CD,),OH'

-t

H,O

(17)

(18)

to reactions 3-6 observed for methanol. Again it is observed that no deuterium is incorporated into the ionic product of reaction 18 indicating that the symmetric proton transfer process, reaction 17, is fast compared to the condensation reaction. The rate constants for reactions 17 and 18 were thus taken from the limiting slope for the disappearance of the deuterated and protonated ions, respectively. Rate data for the labeled and unlabeled ethanol are summarized in Table I.

Discussion 1. Deexcitation Mechanisms. In order to meaningfully compare theoretical rate constants to the experimentally measured rate of conversion of deuterated to protonated parent molecules various statistical corrections must be applied which depend upon the mechanism of the proton transfer reaction. The occurrence of fast, symmetric proton transfer and the formation of collisionally stabilized proton bound dimers at higher pressures are convincing evidence for an excited proton bound dimer as the intermediate in the reactions of protonated alcohols and amines with their parent neutrals. Such an intermediate formed in the interaction of the deuterated alcohol, ROHD', with the parent alcohol, ROH, may take the form

\

H R -0 D'

R-0

H

I \

I \

D

H' H

I1

dimer, 11, with equal probability ignoring any deuterium isotope effect. Dissociation of intermediate I will never lead to the isotopically exchanged ionic product ROH2+. However intermediate I1 has equal probabilities of dissociating to yield ROH2' and ROHD'. Thus only onefourth of the total encounters of ROHD' with ROH will lead to observable reaction. In order to obtain the total encounter rate then, the observed rate of isotopic exchange must be multiplied by a factor of 4. A similar consideration of the possible intermediates formed between CD3NH2D+and CD3NH2and between (CDJ2NHD+and (CD3)2NHleads to statistical correction factors of three in the methylamine case and four in the dimethylamine case. The statistically corrected proton transfer rate constants for the systems studied here are shown in Table I1 together with the total rate of reaction of the parent ions and the encounter rate constants predicted on the basis of the Langevin model'' and the ADO theory.lg Also included for comparison are the rates for similar reactions in ammonia and trimethylamine which have been reported elsewhere.'0'20The summarized data show generally good agreement with rate constants calculated from the ADO theory, indicating that a proton bound dimer intermediate is likely formed in almost every ion-neutral encounter. This conclusion has important implications for the deexcitation of protonated alcohols and amines formed with appreciable amounts of internal or translational energy. If the lifetime of the proton bound dimer intermediate is long compared to the time necessary for redistribution of energy throughout the complex then roughly one-half of the excess energy will appear in both the ionic and neutral products of reaction. If, however, the lifetime of the intermediate is short enough that no energy flows from the excited to the thermal part of the complex during the proton transfer event then there is a probability of one-half that all of the excess energy will remain in the ionic product. In this latter case one-half of the ion-molecule encounters lead to complete thermalization of the ions while in the former case each collision results in loss of one-half of the excess energy of the The Journal of Physical Chemlstv, Vol. 81, No. 7, 1977

598

ion. In both cases it may be assumed that even if the original excitation of the protonated molecule is as much as 100 kcal/mol after as few as 10 collisions the system will be thermalized within the detectable limit of 0.1 kcal/mol. This result is extremely important in considering the validity of thermochemical measurements made from observation of gas phase proton transfer equilibria. Assuming an encounter rate of 10-~ cm3 molecule-’ s-l in trapped ion ICR experiments at lo4 Torr each ion will undergo more than six collisions in a total reaction time of 200 ms which will be more than sufficient to thermalize an ion with 3 kcal/mol excess energy. Since 3 kcal/mol is the maximum free energy difference which may be conveniently measured in ICR observations of bimolecular equilibria it is fairly certain that under typical experimental conditions all ions will have been completely thermalized and a true thermodynamic equilibrium achieved. It has been for this reason that generally good agreement has been obtained among the various diverse techniques for observing ion-molecule equilibria. It should be noted that in cases where symmetric proton transfer is very slow it will take a much higher number of collisions to thermalize ions and the time necessary to achieve thermodynamic equilibrium will be greatly extended. 2. Rate Constants from ICR Experiments. An interesting and extremely important aspect of the results presented here is the good agreement obtained between rate constants determined using the Marshall-Buttrill analysis of intensity vs. pressure data and those determined using the ICR trapped-ion technique. In addition to providing corroboration of rate determinations the two techniques frequently provide complementary results. For example, in the case of ethanol where a secondary ion is produced by more than one primary ion no rate constants may be readily obtained from intensity vs. pressure data. However using trapped-ion techniques rate constants for reaction of each of the primary ions are easily determined. Conversely, when a single primary ion reacts to give two or more reactive secondary ions it is difficult to obtain individual rate constants for production of each secondary ion from a simple analysis of the trapped-ion data. However using the Marshall-Buttrill analysis of intensity vs. pressure data the individual rate constants are all obtained. From the above observations it is therefore highly desirable to be able to carry out both types of ICR experiment on a routine basis to completely characterize the kinetics of moderately complex systems. The advantage of the spectrometer such as the one used in these studies which is able to carry out both experiments facilely is thus evident.

The Journal of Physical Chemistry, Vol. 81, No. 7, 1977

T. B. McMahon and J. L. Beauchamp

Conclusion The results summarized above show that ICR experiments involving appropriately isotopically labeled compounds may be conveniently used to determine rates of symmetric proton transfer reactions. These reactions have been demonstrated to be efficient in the relaxation of epithermal ion energy distributions and hence the symmetric proton transfer rate constant provides a measure of the conditions necessary for thermalization of ionic reactants. This knowledge is extremely useful in assessing the validity of thermochemical inferences made from observation of ion-molecule equilibria. Acknowledgment. This work was supported in part by the Energy Research and Development Administration under Grant No. AT(04-3)767-8. T. B. McMahon thanks the National Research Council of Canada for fellowship support.

References and Notes (1) Present address: Department of Chemistry, University of New Brunswick. Fredericton, New Brunswick. Canada. (2) (a) Camille and Henry Dreyfus Teacher-Scholar, 1971-1976. (b) Contribution No. 5264. (3) M. A. Haney and J. L. Franklin, J. Chem. Phys., 48, 4093 (1966). (4) R. H. Staley and J. L. Beauchamp, J. Chem Phys, 62, 1998 (1975). (5) J. P. Briggs, R. Yamdagni, and P. Kebarle, J. Am. Chem. SOC.,94, 5128 (1972); D. H. Aue, H. M. Webb, and M. T. Bowers, ibid., 94, 4726 (1972); W. G. Henderson, M. Taagepera, D. Hob. R. T. McIver, Jr., J. L. Beauchamp and R. W. Taft, ibid., 94, 4728 (1972). (6) See,for example, J. J. Leventhaland L. Friedman, J. Chem. Phys, 50, 2928 (1969); M. L. Vestal, C. R. Blakley, P. W. Ryan, and J. H. Futrell, ibid., 64, 2094 (1976). (7) J. L. Beauchamp in “Interactions between Ions and Molecules”, P. Ausloos, Ed., Plenum Press, New York, N.Y., 1975, pp 413-444 and references contained therein. (8) L. W. Sieck, F. P. Abramson, and J. H. Futrell, J. Chem. Phys., 45, 1655 (1966). (9) M. S. B. Munson and F. H. Field. J. Am Chem Soc..87. 4242 (1965). (i0i A. G. Harrison, P. H. Lin, and C. W. Tsang, Inf. J. Mass Specfroh. Ion Phys., 19, 23 (1976). (11) T. B. McMahon, P. G. Miasek, and J. L. Beauchamp, Inf. J. Mass Specfrom. Ion Phys., 21, 63 (1976). (12) A. G. Marshall and S. E. Buttrill, Jr., J. Chem Phys, 52, 2752 (1970). (13) T. B. McMahon and J. L. Beauchamp, Rev. Sci. Insfrum., 43, 509 (1972). (14) G. J. Shulz and A. V. Phelps, Rev. Sci. Instrum., 28, 1051 (1957). (15) K. R. Ryan, L. W. Sieck, and J. H. Futrell, J. Chem. Phys., 41, 111 (1964). (16) S. K. Gupta, E. G. Jones, A. G. Harrison, and J. J. Myher, Can. J. Chem., 45, 3107 (1967). (17) J. M. S. Henis, J. Am. Chem. Soc., 90, 844 (1968). (18) (a) P. Langevin, Ann. Chim Phys., 5 , 245 (1905); (b) G. Gioumousiss and D. P. Stevenson, J. Chem. Phys., 29, 294 (1958). (19) T. Su and M. T. Bowers, J. Chem. Phys., 50, 3027 (1973). (20) T. B. McMahon, Ph.D. Thesis, California Instituteof Technology, 1973. The rate constant for symmetric proton transfer involving (CD,),N is estimated as the average of the forward and reverse rate constants for the proton transfer equilibrium (CD,),NH+ f (CH,)3N ==(CD&,N (CH&NH+.

+